Recently, mass spectrometry chemists have implemented in-silico machine learning algorithms to analyze their data for molecular determination. As of now, open-access models that exist perform on either liquid chromatography-mass spectrometry (LC-MS) data, or low resolution gas chromatography-mass spectrometry (GC-MS) electron ionization (EI) data. Since LC-MS is a relatively recent technique, it consistently delivers high-resolution data. In contrast, GC-MS, a hard ionization technique with significant fragmentation, traditionally yielded low-resolution data since it was developed in the 1950s. Over the past two decades, enhancements to GC-MS have yielded high-resolution data. Consequently, there is a need to develop

high-resolution GC-MS spectrum libraries to initiate the machine learning process

for such data. With our own instrumentation and publicly available datasets, we can gather millions of spectral information to construct a local database to train the algorithm on high resolution GC-MS data.